基于复Contourlet域非线性扩散的图像去噪  

Image de-noising based on complex contourlet transform and nonlinear diffusion

作　　者：吴一全[1,2] 侯雯[1] 吴诗婳[1] 

机构地区：[1]南京航空航天大学电子信息工程学院,江苏南京210016 [2]光电控制技术重点实验室,河南洛阳471009

出　　处：《电路与系统学报》2012年第6期111-116,共6页Journal of Circuits and Systems

基　　金：国家自然科学基金资助项目(60872065);光电控制技术重点实验室和航空科学基金联合资助项目(20105152026);南京大学计算机软件新技术国家重点实验室开放基金资助项目(KFKT2010B17)

摘　　要：提出了一种将复Contourlet变换与非线性扩散相结合的图像去噪方法。首先对图像进行复Contourlet分解,然后高频部分和低频部分分别采用自适应对比度扩散和全变差扩散,最后重构图像。给出了实验结果,并与基于小波、基于Contourlet和基于非下采样Contourlet的非线性扩散方法的图像去噪效果进行了主观视觉上的比较,同时也依据均方差(MSE)、峰值信噪比(PSNR)等评价指标作了定量分析,且对比了各算法的运行时间。结果表明,本文提出的方法去噪效果更为优越:不但抑制噪声的能力更强,而且能够更好地保留图像原有的边缘和纹理特征。An image de-noising method combining complex contourlet transform with nonlinear diffusion is proposed in this paper. Firstly, an image is decomposed by complex contourlet transform. Then adaptive-contrast-factor diffusion and total variation diffusion are applied to high-frequency component and low-frequency component, respectively. Finally the image is synthesized. The experimental results are given. Subjective-visual-quality comparisons of the image de-noising results are made with those of the image de-noising methods based on the combination of wavelet/contourlet/nonsubsampled contourlet with nonlinear diffusion, while the experimental results are evaluated quantitatively according to the evaluation items such as mean square error（MSE）, peak signal to noise ratio（PSNR） and so on. Running time of the above-mentioned algorithms is also contrasted. It is shown that the proposed image de-noising methods based on complex contourlet transform and nonlinear diffusion can obtain superior results, which can both remove noise and preserve the original edges and textural features more efficiently.

关 键 词：图像去噪 复Contourlet变换 全变差扩散 自适应对比度扩散 

分 类 号：TN911.73[电子电信—通信与信息系统]